SUMMARY Interactions between proteins play crucial roles in biology. Over- or under-expression of specific proteins can lead to aberrant interaction levels that contribute to disease. In addition, many pathogens depend on protein-protein interactions for infection. Thus, developing tools that lead to new inhibitors of specific protein-protein interactions, or molecules that can substitute for a missing or under- expressed partner, is a goal of considerable importance. Such molecules can serve as powerful tools for elucidating the biological functions of particular interactions, and they can provide the basis for new therapeutic agents. The proposed research is intended to generate new strategies for creating molecules that bind tightly to recognition surfaces displayed by natural proteins, surfaces that evolved to form specific contacts with other proteins or large peptides. The compounds we seek could disrupt deleterious protein associations or boost inadequate interaction levels. Currrent approaches to these goals are based on small molecules, engineered proteins or medium-sized conventional peptides (i.e., peptides comprised exclusively of -amino acid residues). Although these approaches can be successful, each has limitations. For example, small molecules often cannot cover enough surface area on a given protein for effective inhibition of association with a large partner protein. Proteins engineered for therapeutic applications can be expensive to produce and store, and longterm use can provoke a deleterious immune response. Medium-sized conventional peptides are often degraded rapidly by proteases in vivo. Our approach is intended to complement these existing strategies for generating molecules that bind to specific surfaces on target proteins. We focus on oligomers that contain both - and -amino acid residue ( / -peptides). Our recent work shows that / -peptides containing 25- 33% residues interspersed evenly among the residues can be highly resistant to proteolysis. In addition, we have found that residues with a specific cyclic constraint can strongly stabilize an - helix-like conformation. One aspect of the proposed research involves a search for alternative residue constraints that match non-helical local conformations commonly adopted by -amino acid residues in folded proteins. This goal is being pursued via a combination of experimental and computational methods, in the context of inhibiting the association of a soluble signaling protein with its cell-surface receptors. Another aspect of the proposed research focuses on new agonists for B- family GPCRs. These studies allow us to determine whether / -peptide analogues of natural agonists can display functional selectivity in their interactions with the target receptors (biased agonism). The over-arching goal of this program is to develop broadly applicable design strategies that can be implemented in many laboratories and that will be useful for many biomedically important protein-protein interactions.